RT Journal Article
T1 Edgeworth expansions for spectral density estimates and studentized sample mean
A1 Velasco, Carlos
A1 Robinson, Peter M.
AB We establish valid Edgeworth expansions for the distribution of smoothed nonparametricspectral estimates, and of studentized versions of linear statistics such as the sample mean, where the studentization employs such a nonparametric spectralestimate. Particular attention is paid to the spectral estimate at zero frequencyand, correspondingly, the studentized sample mean, to reflect econometric interestin autocorrelation-consistent or long-run variance estimation. Our main focusis on stationary Gaussian series, though we discuss relaxation of the Gaussianityassumption. Only smoothness conditions on the spectral density that are local tothe frequency of interest are imposed. We deduce empirical expansions from ourEdgeworth expansions designed to improve on the normal approximation in practiceand also deduce a feasible rule of bandwidth choice.
PB Cambridge University Press
SN 1469-4360
YR 2001
FD 2001
LK https://hdl.handle.net/10016/3970
UL https://hdl.handle.net/10016/3970
LA eng
DS e-Archivo
RD 19 may. 2024